Newsheise Security
OpenAI führt Framework zur Meldung von KI-Sicherheitsvorfällen ein
Teaser from the source
OpenAI hat ein neues Framework vorgestellt, um Fehlverhalten seiner KI-Modelle systematisch zu erfassen, zu untersuchen und transparent offenzulegen.
This is the RSS-feed teaser. Read the full article at the original source.
Read at heise SecurityMore on OpenAI
Other advisories
- CriticalCVE-2026-119990Missing authentication for critical function in Azure OpenAI allows an unauthorized attacker to elevate privileges over a network.
- CriticalCVE-2026-119990Azure OpenAI Elevation of Privilege Vulnerability
- Not ratedosv:openSUSE-SU-2026:12488-1openai-codex-0.166.0-2.1 on GA media
- HighCVE-2026-110930OpenAI Codex Desktop for Windows: Argument-Injection
- HighCVE-2026-110930OpenAI Codex Desktop for Windows: Argument-Injection
- Not ratedosv:CVE-2026-121936easyVoice before 0.0.23 SSRF via openaiBaseUrl in TTS API
- Not ratedosv:CLEANSTART-2026-AQ37229LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
- Not ratedosv:CLEANSTART-2026-HA67725LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
Other news entries
- OpenAI launches security products, tells orgs they’re doomed if they don’t use themNewsThe Register — Security
- Nach Sicherheitsvorfall mit OpenAI: Australien prüft Straftatbestand gegen unsichere KINewsGolem.de — Security
- OpenAI builds a kill switch for its research modelsNewsThe Register — Security
- Former OpenAI Engineer Pleads Not Guilty to Planting Covert Device in Executive Meeting RoomNewsThe Hacker News
- OpenAI's new GPT-6 Cyber AI finds 81 Firefox flaws, 948 Linux bugsNewsBleepingComputer
- OpenAI Launches GPT-6 Cyber After AI Helps Find 81 Firefox Flaws, 948 Linux ExploitsNewsThe Hacker News
- OpenAI Says New GPT-6 Cyber Model Hunts Flaws Without Crossing the LineNewsSecurityWeek
- OpenAI Model Wrote Its Own Network Code to Get Around Sandbox RestrictionsNewsSecurityWeek
ID